As one of structures of a solar cell having relatively high photoelectric conversion efficiency using a single-crystal or polycrystalline semiconductor substrate, there is a back surface electrode type solar cell in which all of positive electrodes and negative electrodes are provided on a non-light receiving surface (back surface). FIG. 1 illustrates an overview of the back surface of the back surface electrode type solar cell. Emitter layers 102 and base layers 101 are alternatively arranged, and electrodes (103, 104) are provided on the layers along the respective layers. The emitter layer has a width of several mm to several hundred μm, and the base layer has a width of several hundred μm to several ten μm. Further, the electrode typically has a width of approximately several hundred to several ten μm, and the electrode is often referred to as a finger electrode.
FIG. 9 illustrates a schematic diagram of a cross-sectional structure of the back surface electrode type solar cell. In the vicinity of an outermost layer of a back surface of a substrate 202, the emitter lavers 102 and the base layers 101 are formed. The thickness of each layer is at most approximately 1 μm. The finger electrodes (205, 206) are provided on the respective layers, and the surface of a non-electrode region is covered with a back surface protective coat 207 such as a silicon nitride film and a silicon oxide film. An antireflective coat 201 is provided on a light receiving surface side for the purpose of reducing reflectance loss.